{"ID":2829278,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.12617","arxiv_id":"2512.12617","title":"Spectral Sentinel: Scalable Byzantine-Robust Decentralized Federated Learning via Sketched Random Matrix Theory on Blockchain","abstract":"Decentralized federated learning (DFL) enables collaborative model training without centralized trust, but it remains vulnerable to Byzantine clients that poison gradients under heterogeneous (Non-IID) data. Existing defenses face a scalability trilemma: distance-based filtering (e.g., Krum) can reject legitimate Non-IID updates, geometric-median methods incur prohibitive $O(n^2 d)$ cost, and many certified defenses are evaluated only on models below 100M parameters. We propose Spectral Sentinel, a Byzantine detection and aggregation framework that leverages a random-matrix-theoretic signature: honest Non-IID gradients produce covariance eigenspectra whose bulk follows the Marchenko-Pastur law, while Byzantine perturbations induce detectable tail anomalies. Our algorithm combines Frequent Directions sketching with data-dependent MP tracking, enabling detection on models up to 1.5B parameters using $O(k^2)$ memory with $k \\ll d$. Under a $(σ,f)$ threat model with coordinate-wise honest variance bounded by $σ^2$ and $f \u003c 1/2$ adversaries, we prove $(ε,δ)$-Byzantine resilience with convergence rate $O(σf / \\sqrt{T} + f^2 / T)$, and we provide a matching information-theoretic lower bound $Ω(σf / \\sqrt{T})$, establishing minimax optimality. We implement the full system with blockchain integration on Polygon networks and validate it across 144 attack-aggregator configurations, achieving 78.4 percent average accuracy versus 48-63 percent for baseline methods.","short_abstract":"Decentralized federated learning (DFL) enables collaborative model training without centralized trust, but it remains vulnerable to Byzantine clients that poison gradients under heterogeneous (Non-IID) data. Existing defenses face a scalability trilemma: distance-based filtering (e.g., Krum) can reject legitimate Non-I...","url_abs":"https://arxiv.org/abs/2512.12617","url_pdf":"https://arxiv.org/pdf/2512.12617v1","authors":"[\"Animesh Mishra\"]","published":"2025-12-14T09:43:03Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.DC\"]","methods":"[]","has_code":false}